Impaired consciousness during epileptic seizures has serious consequences, including hazardous driving, decreased work/school performance and social stigmatization. Patients with medically and surgically refractory epilepsy often suffer from partial seizures with impaired consciousness, and no viable treatment options. Recent neuroimaging and electrophysiology findings suggest that impaired consciousness in partial seizures is related to depressed cortical function in widespread regions remote from the seizure focus. We have developed a rat model that replicates the human findings, including slow waves on electro- encephalography (EEG), decreased cerebral blood flow (CBF) and decreased functional MRI (fMRI) signals in the neocortex, as well as behavioral arrest during seizures. Advances based on both human work and the rat model demonstrates that focal seizures decrease activity in subcortical arousal structures including the upper brainstem and intralaminar thalamus. This in turn leads to sleep- or coma-like changes in the association cortex, and to loss of consciousness during and following seizures. In other disorders of consciousness, such as minimally conscious state, recent work has shown that thalamic stimulation can increase behavioral arousal. These finding pave the way for a new therapeutic approach to directly treat impaired consciousness in partial seizures: reactivation of the subcortical arousal systems through deep brain stimulation. The goal of the current proposal is to test this approach in the rat model, providing proof-of-principle efficacy for possible human therapeutic trials. Our preliminary studies suggest that stimulation of the rostral thalamic intralaminar centro-lateral nucleus (CL) converts slow wave activity in the cortex into an awake EEG pattern, and also increases cortical function based on fMRI. Therefore our aims are to first test CL thalamic stimulation under general anesthesia, to emulate the conditions during surgical device implantation. We will obtain suitable therapeutic stimulus parameters aimed to achieve cortical activation based on electrophysiology and intraoperative fMRI. Second, we will test the therapeutic efficacy of CL stimulation to improve cortical function and behavioral responsiveness during seizures in awake, behaving animals. We will test the effects of stimulation during both evoked seizures, and in response to spontaneous seizures-automatically detected using closed-loop stimulation similar to that employed in human devices. Although stopping seizures is ideal, for patients in whom seizures cannot be stopped, preventing impaired consciousness would greatly improve quality of life. If deep brain stimulation can improve cortical function and consciousness during partial seizures in an animal model, this may rapidly lead to translation of this new approach to human treatment trials.